Wheel Torque Control Using RBF Networks for Faster Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle dynamics control systems require extensive calibration for each new vehicle type and road condition, which is time-consuming and inefficient.
Innovation Solution
A method using a radial basis function (RBF) network to determine torque changes for wheel control, incorporating current slip, acceleration, and historical force and torque data, allowing for automated and robust torque calibration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional vehicle dynamics control systems are used with extensive calibration for each new vehicle type, then control accuracy is improved, but calibration time and complexity increase significantly
Solution Approach 1:
The patent transforms the calibration problem from a time-consuming manual parameter adjustment process into an automated learning process. The neural network automatically learns optimal control parameters from training data, eliminating the need for extensive manual calibration while maintaining or improving control accuracy. This resolves the contradiction by changing the fundamental approach from manual parameter tuning to automated parameter learning.
Solution Approach 2:
The patent replaces the manual mechanical calibration process with an automated computational system. Instead of physically adjusting components and manually testing each vehicle type, the system uses a neural network that can be trained once and then automatically applied to multiple vehicle types, dramatically reducing calibration time while maintaining control accuracy.
2Adaptability or versatility
If traditional control systems are manually calibrated for different road conditions, then adaptability is improved, but system complexity and calibration effort increase
Solution Approach 1:
The patent creates a universal control system where a single neural network model can adapt to multiple vehicle types and road conditions through training. The same trained network can be applied across different platforms without requiring separate calibration for each scenario, thereby improving adaptability while reducing overall system complexity.
Solution Approach 2:
The patent performs the adaptation work in advance during the training phase. The neural network is pre-trained on comprehensive datasets covering various vehicle types and road conditions, so that during actual operation, it can immediately adapt to new conditions without requiring complex real-time calibration procedures.
3Manufacturing precision
If extensive calibration is performed for each vehicle type, then control precision is improved, but productivity and time to market decrease
Solution Approach 1:
The patent replaces the sequential manual calibration process with a parallel automated training process. The neural network can be trained on multiple vehicle types simultaneously using computational resources, dramatically increasing productivity and reducing time to market while maintaining control precision through the network's ability to learn optimal parameters for each vehicle type.
Solution Approach 2:
The patent performs all necessary calibration work in advance during the development phase. Once the neural network is trained, it can be deployed across multiple vehicle types without requiring additional calibration time, thereby accelerating productivity and reducing time to market while preserving control precision.
Data Source
Figure 1
AI summary
The invention relates to a method for controlling a torque of at least one wheel of a mobile platform, comprising the following steps: - providing at least one current slip value of the wheel and at least one current wheel acceleration of the wheel as input values; - providing a trained radial basis function network designed to determine, by means of the input values, at least one torque change as an output value for control of the at least one wheel; and - determining a current torque change, by means of the trained radial basis function network and the provided input values, for control of the torque.